基于全景视觉机器人自定位方法研究  

The Study of Self-localization Method Based on Omni-direction Vision

作　　者：邓本再[1] 黄苗[1] 谢志诚[1] 李亘[1] 张飞[1] 

机构地区：[1]长沙理工大学.电气与信息工程学院,湖南长沙410114

出　　处：《计算技术与自动化》2009年第4期99-102,共4页Computing Technology and Automation

摘　　要：根据RoboCup比赛规则的改变,在足球机器人失去利用球门、立柱等地标信息定位条件情况下,提出一种新的基于全景视觉的自定位方法。该方法通过对白色标志线上点的识别,结合场地环境栅格化的处理,通过最小误差搜索而达到快速、精确自定位的目的。实验结果表明该自定位方法达到了较高定位精度和较快定位速度。According to the change of the rule of middle--size league of RoboCup, soccer robot lost the available Land-mark information of Column and goal. A new self --localization algorithm is proposed In this paper. This self--localization algorithm is based on the white points of mark lines;combining with gird of Space environment, by searching the smallest error, thus, reach to the goal of faster and precision self--localization. The results show that the algorithm with high precision and faster location.

关 键 词：机器人 全景视觉 自定位 最小误差 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]